JP7628478B2 - Wire saw abnormality diagnosis device and method - Google Patents

Description

The present invention relates to a wire saw abnormality diagnosis device and method.

There are wire saws that use wire to cut workpieces such as semiconductor materials and magnetic materials (see, for example, Patent Document 1). This wire saw cuts the workpiece by running a wire wound around multiple processing rollers at a predetermined interval at high speed and pressing the workpiece against the wire. Wire saws generally include multiple processing rollers around which the wire is wound, a pay-out reel that pays out the wire toward the processing rollers, and a take-up reel that winds up the wire sent from the processing rollers.

The device described in Patent Document 1 measures the tension of the wire and the vibration using a vibration sensor attached to the workpiece holder before cutting the workpiece, and uses these measurements to determine whether the wire saw is normal or abnormal. If the wire saw is determined to be normal, it cuts the workpiece, thereby preventing deterioration in the quality of products such as wafers cut from the workpiece (see paragraphs [0009] and [0010] of Patent Document 1).

Patent No. 6729697

In the device of Patent Document 1, it is believed that the operation of the wire saw during the measurement performed before cutting the workpiece is performed under the same conditions as the operation during the subsequent cutting process, for example, the wire travel speed (linear speed) is constant.

However, wire saws have a complex configuration that requires multi-axis synchronous control of the processing roller, the payout reel, and the take-up reel, and when operating under constant linear speed conditions, the amount of wire stored and the effective diameter of the payout reel and the take-up reel change each time. Furthermore, during maintenance, repairs may be performed by grinding the grooves on the outer peripheral surface of the processing roller, which may cause changes that reduce the effective diameter. For this reason, the device in Patent Document 1 does not adequately monitor the operating state before cutting, making it difficult to accurately diagnose abnormalities.

The objective of the present invention is to thoroughly monitor the operating state of a wire saw before cutting and accurately diagnose abnormalities.

In order to solve the above problem, the present invention is an abnormality diagnosis device for a wire saw that cuts a workpiece with a wire. The wire saw includes a plurality of processing rollers, a wire wound around the plurality of processing rollers, a payout reel that pays out the wire toward the processing rollers, and a take-up reel that takes up the wire fed from the processing rollers. The abnormality diagnosis device includes a diagnostic mode execution unit that executes a first diagnostic mode in which the wire saw is operated so that the processing rollers rotate at a constant first rotation speed before the cutting process, a second diagnostic mode in which the wire saw is operated so that the payout reel rotates at a constant second rotation speed, and a third diagnostic mode in which the wire saw is operated so that the take-up reel rotates at a constant third rotation speed, and a diagnostic mode execution unit that executes a first data group obtained in the first diagnostic mode, a second data group obtained in the second diagnostic mode, and a third data group obtained in the third diagnostic mode for a plurality of data items indicating the operating state of the wire saw. a data group acquisition unit that acquires each of the data groups; a deviation information calculation unit that calculates deviation information regarding deviations obtained by comparing a first reference data group obtained in the first diagnostic mode when the wire saw is normal, a second reference data group obtained in the second diagnostic mode when the wire saw is normal, and a third reference data group obtained in the third diagnostic mode when the wire saw is normal, with the first data group, the second data group, and the third data group, respectively, for a plurality of the data items; and a determination unit that determines the presence or absence of an abnormality in the wire saw based on the calculated deviation information.

The present invention also relates to a method for diagnosing an abnormality in a wire saw that cuts a workpiece with a wire. The abnormality diagnosis method includes the steps of: before the cutting process, executing a first diagnostic mode in which the wire saw is operated so that the processing roller rotates at a constant first rotation speed; a second diagnostic mode in which the wire saw is operated so that the pay-out reel rotates at a constant second rotation speed; and a third diagnostic mode in which the wire saw is operated so that the take-up reel rotates at a constant third rotation speed; and, for a plurality of data items indicating the operating state of the wire saw, executing a first data group obtained in the first diagnostic mode, a second data group obtained in the second diagnostic mode, and a third data group obtained in the third diagnostic mode. The method includes a step of acquiring three data groups, a step of calculating deviation information for a plurality of the data items, the deviation of which is obtained by comparing a first reference data group obtained in the first diagnostic mode when the wire saw is normal, a second reference data group obtained in the second diagnostic mode when the wire saw is normal, and a third reference data group obtained in the third diagnostic mode when the wire saw is normal with the first data group, the second data group, and the third data group, respectively, and a step of determining whether or not there is an abnormality in the wire saw based on the calculated deviation information.

According to the present invention, the operating state of the wire saw before cutting can be adequately monitored and abnormality diagnosis can be performed accurately.

1 is a diagram showing a schematic configuration of a wire saw to which an abnormality diagnosis device according to an embodiment of the present invention is applied; FIG. 2 is a schematic block diagram showing a control configuration of the wire saw. 4 is a functional block diagram showing an internal configuration related to abnormality diagnosis in the control device of the wire saw. FIG. 4 is a flowchart showing the contents of an abnormality diagnosis method according to the present embodiment.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail with reference to the accompanying drawings.
In each drawing, common or similar components are denoted by the same reference numerals, and their repeated explanations are omitted as appropriate. Furthermore, the size and shape of the components may be exaggerated or distorted for ease of explanation.

FIG. 1 is a diagram showing a schematic configuration of a wire saw 1 to which an abnormality diagnostic device according to an embodiment of the present invention is applied.
1, the wire saw 1 cuts a workpiece 15 with a wire 2. The wire saw 1 includes a plurality of processing rollers 6, a payout reel 3 that pays out the wire 2 toward the processing rollers 6, and a take-up reel 9 that takes up the wire 2 fed from the processing rollers 6. For example, a fixed-abrasive wire 2 is wound around the plurality of processing rollers 6 that extend in parallel at predetermined intervals.

A number of annular grooves (not shown) are formed at a predetermined pitch on the outer circumferential surface of the processing roller 6, and a wire 2 made of a single wire rod is wound continuously around each groove of the processing rollers 6. In this embodiment, two processing rollers 6 are provided.

The wire 2 is unwound from the unwinding reel 3, passes through the unwinding traverse device 4 and the dancer roller device 5, is wound around two processing rollers 6 in multiple layers, and passes through the winding dancer roller device 7 and the traverse device 8 before being wound onto the winding reel 9.

The wire 2 travels back and forth between two processing rollers 6. In this case, the wire 2 alternates between moving forward a fixed amount and then moving backward a fixed amount that is smaller than the fixed amount of forward movement, and is driven to move forward stepwise overall. Note that the wire 2 may also be driven to move forward continuously in one direction.

The traverse devices 4 and 8 are provided to always orthogonally orthogonally align the wire 2 with the axis of the payout reel 3 or the take-up reel 9, respectively. The dancer roller devices 5 and 7 are provided to apply a constant tension to the wire 2 while allowing the wire 2 to travel back and forth by the reciprocating motion of the dancer rollers (not shown).

The wire 2 forms a cutting area in the area between the two processing rollers 6 facing the workpiece 15 to be processed. The spacing of the wire 2 in the cutting area is regulated by the pitch of the grooves formed on the outer circumferential surface of the processing rollers 6. This groove pitch determines the thickness of the plate-like body, i.e., wafer, after the workpiece 15 has been cut.

The supply reel 3, the processing roller 6, and the take-up reel 9 are rotated by a motor 11 for driving the supply reel, a motor 12 for driving the processing roller, and a motor 13 for driving the take-up reel, respectively. All of these motors 11, 12, and 13 are controlled by a control device 50 (see FIG. 2).

The workpiece 15 is fixed to the feed device 16 for processing feed at a position corresponding to the cutting area of the wire 2 between the two processing rollers 6, and is pressed against the cutting area of the wire 2 at a predetermined feed speed. The workpiece feed motor 14 of the feed device 16 has a built-in rotary-linear motion conversion mechanism (not shown), such as a feed screw unit. The workpiece feed motor 14 is under the control of the control device 50, and moves the workpiece 15 in the processing feed direction at a predetermined feed speed according to the feed movement amount of the workpiece 15 for cutting processing of the workpiece 15. The feed movement amount corresponds to the feed position or cutting depth of the workpiece 15, and is detected by a signal from an encoder 20 connected to the workpiece feed motor 14.

Encoders 17, 18, and 19 are connected to motor 11 for driving the supply reel, motor 12 for driving the processing roller, and motor 13 for driving the take-up reel, respectively. Therefore, the rotational speeds of supply reel 3, processing roller 6, and take-up reel 9 are detected by signals from encoders 17, 18, and 19, respectively.

FIG. 2 is a schematic block diagram showing a control configuration of the wire saw 1.
2, the wire saw 1 includes a control device 50 that controls each part of the wire saw 1. The control device 50 includes a CPU (Central Processing Unit) (not shown) and a storage unit such as a RAM, a ROM, and a hard disk. A program for executing the abnormality diagnosis method according to this embodiment is stored in the storage unit.

The control device 50 receives signals from the encoders 17-20. The control device 50 controls the driving of the motors 11-14 via the drivers 21-24, respectively. The control device 50 is also connected to an input device 25 and a display device 26. The input device 25 receives input of various information such as user operations. The display device 26 displays various information such as an operation screen, a display screen, and warning messages. The control device 50 also receives signals from the vibration detection means 31-33 and the temperature detection means 34-36.

The vibration detection means 31 detects the vibration characteristics of the pay-out reel 3 or the support member of the pay-out reel 3 (not shown). The vibration detection means 32 detects the vibration characteristics of the processing roller 6 or the support member of the processing roller 6 (not shown). The vibration detection means 33 detects the vibration characteristics of the take-up reel 9 or the support member of the take-up reel 9 (not shown). The support members of the processing roller 6, the pay-out reel 3, and the take-up reel 9 include, for example, bearings and casings in which the bearings are installed. The temperature detection means 34 detects the temperature of the pay-out reel 3 or the support member of the pay-out reel 3. The temperature detection means 35 detects the temperature of the processing roller 6 or the support member of the processing roller 6. The temperature detection means 36 detects the temperature of the take-up reel 9 or the support member of the take-up reel 9. The vibration detection means 32 and the temperature detection means 35 are provided corresponding to one or both of the two processing rollers 6.

In the wire saw 1, for example, the bearings of the processing roller 6 usually rotate at high speed under high load, and therefore have a short lifespan. In addition, the motor load may become excessively high when driving the processing roller 6 to rotate. In addition, thermal displacement caused by an increase in temperature around the processing area of the workpiece 15 may cause warping of the wafer obtained by cutting the workpiece 15, and the processing quality of the workpiece 15 may decrease. In this way, various abnormalities may occur in the wire saw 1. Even if an abnormality is detected during cutting by the wire saw 1, it may be difficult to repair it immediately, and in such a case, a decrease in yield is unavoidable. In contrast, in this embodiment, an abnormality diagnosis of the wire saw 1 is performed before cutting.

FIG. 3 is a functional block diagram showing an internal configuration related to abnormality diagnosis in the control device 50 of the wire saw 1.
As shown in Fig. 3, the control device 50 of the wire saw 1 includes a diagnosis mode execution unit 51, a data group acquisition unit 52, a reference data group setting unit 53, a deviation information calculation unit 54, a judgment unit 55, and a factor estimation unit 56. The units shown in Fig. 3 constitute an abnormality diagnosis device for the wire saw 1, and the functions of these units are realized by executing a predetermined program stored in a memory unit in the control device 50 on a RAM. Note that each unit shown in Fig. 3 may be a control circuit configured by hardware.

The diagnostic mode execution unit 51 executes a first diagnostic mode, a second diagnostic mode, and a third diagnostic mode. The first diagnostic mode is a mode in which the wire saw 1 is operated so that the processing roller 6 rotates at a constant first rotation speed before cutting. The second diagnostic mode is a mode in which the wire saw 1 is operated so that the pay-out reel 3 rotates at a constant second rotation speed before cutting. The third diagnostic mode is a mode in which the wire saw 1 is operated so that the take-up reel 9 rotates at a constant third rotation speed. In these first diagnostic mode, second diagnostic mode, and third diagnostic mode, operation is performed in a state in which the wire 2 is not in contact with the workpiece 15 (hereinafter also referred to as "idle operation").

The data group acquisition unit 52 acquires a first data group obtained in a first diagnostic mode, a second data group obtained in a second diagnostic mode, and a third data group obtained in a third diagnostic mode. The first to third data groups are obtained for multiple data items that indicate the operating state of the wire saw 1.

In this embodiment, the multiple data items when obtaining the first data group are the torque load of the motor 12 that rotates the processing roller 6, the temperature of the processing roller 6 or the support member (not shown) of the processing roller 6, and the vibration characteristics of the processing roller 6 or the support member of the processing roller 6. However, other data items may be included. The multiple data items when obtaining the second data group are the torque load of the motor 11 that rotates the pay-out reel 3, the temperature of the pay-out reel 3 or the support member (not shown) of the pay-out reel 3, and the vibration characteristics of the pay-out reel 3 or the support member of the pay-out reel 3. However, other data items may be included. The multiple data items when obtaining the third data group are the torque load of the motor 13 that rotates the take-up reel 9, the temperature of the take-up reel 9 or the support member (not shown) of the take-up reel 9, and the vibration characteristics of the pay-out reel 3 or the support member of the pay-out reel 3. However, other data items may be included.

The vibration characteristics may be amplitude, frequency, or features called the amount of change or the amount of existence. The torque loads of the motors 11 to 13 are obtained as torque command values output from the control device 50 to the drivers 21 to 23, respectively, or as drive current values output from the drivers 21 to 23 to the motors 11 to 13, respectively.

The reference data group setting unit 53 sets a first reference data group, a second reference data group, and a third reference data group, which are data groups constituting a unit space. The first reference data group is obtained in a first diagnostic mode when the wire saw 1 is normal, for the same multiple data items as when the first data group is obtained. The second reference data group is obtained in a second diagnostic mode when the wire saw 1 is normal, for the same multiple data items as when the second data group is obtained. The third reference data group is obtained in a third diagnostic mode when the wire saw 1 is normal, for the same multiple data items as when the third data group is obtained. The first reference data group, the second reference data group, and the third reference data group are set in advance, before the first data group, the second data group, and the third data group are obtained.

The deviation information calculation unit 54 calculates deviation information regarding the deviations obtained by comparing the first, second, and third reference data groups with the first, second, and third data groups, respectively. That is, the deviation information includes first deviation information regarding the deviation between the first reference data group and the first data group, second deviation information regarding the deviation between the second reference data group and the second data group, and third deviation information regarding the deviation between the third reference data group and the third data group. That is, the deviation information includes first deviation information obtained in the first diagnostic mode, second deviation information obtained in the second diagnostic mode, and third deviation information obtained in the third diagnostic mode.

In this embodiment, the first deviation information is the Mahalanobis distance between a first unit space formed by a first reference data group and a first signal space formed by a first data group. The second deviation information is the Mahalanobis distance between a second unit space formed by a second reference data group and a second signal space formed by a second data group. The third deviation information is the Mahalanobis distance between a third unit space formed by a third reference data group and a third signal space formed by a third data group. The deviation information may be configured to be calculated by a deviation information calculation model that has been trained by machine learning.

The determination unit 55 determines whether or not there is an abnormality in the wire saw 1 based on the deviation information calculated by the deviation information calculation unit 54. In this embodiment, this determination is performed by comparing the Mahalanobis distance as the deviation information with a preset threshold value. In other words, the determination unit 55 monitors whether the Mahalanobis distance calculated by the deviation information calculation unit 54 does not exceed a predetermined threshold value.

When the wire saw 1 is operating normally, the Mahalanobis distance falls within a specific range. On the other hand, if the operating state shows signs of being abnormal, the correlation between multiple data items will deviate from the unit space. Therefore, by calculating the Mahalanobis distance used in the MT method and the standardized mispressure method, it is possible to diagnose abnormalities in the operating state of the wire saw 1.

When it is determined that there is an abnormality in the wire saw 1, the cause estimation unit 56 estimates the cause of the abnormality based on the contribution of each of the multiple data items to the Mahalanobis distance as deviation information.

Next, the abnormality diagnosis method according to this embodiment will be described with reference to FIG.
FIG. 4 is a flowchart showing the contents of the abnormality diagnosis method according to this embodiment.
4, the diagnostic mode execution unit 51 executes one of a plurality of diagnostic modes (step S1). Specifically, first, a first diagnostic mode is executed in which the wire saw 1 is idled so that the processing roller 6 rotates at a constant first rotation speed before cutting.

Next, the data group acquisition unit 52 acquires a first data group obtained in the first diagnostic mode for multiple data items indicating the operating state of the wire saw 1 (step S2).

In step S3, it is determined whether or not all of the multiple diagnostic modes, i.e., the first diagnostic mode, the second diagnostic mode, and the third diagnostic mode, have been executed. If all of the multiple diagnostic modes have been executed (Yes in step S3), proceed to step S4. If any of the multiple diagnostic modes remain to be executed (No in step S3), return to step S1.

Steps S1 and S2 are then repeated to obtain a second group of data obtained in a second diagnostic mode in which the wire saw 1 is run idly so that the supply reel 3 rotates at a constant second rotational speed. Also, a third group of data is obtained in a third diagnostic mode in which the wire saw 1 is run idly so that the take-up reel 9 rotates at a constant third rotational speed.

Then, the deviation information calculation unit 54 calculates the deviation information (step S4). The deviation information is information on the deviation obtained by comparing the first reference data group, the second reference data group, and the third reference data group with the first data group, the second data group, and the third data group, respectively. Here, the first to third reference data groups are obtained in advance for the above-mentioned multiple data items in the first to third diagnostic modes when the wire saw 1 is normal, respectively.

In step S5, the determination unit 55 determines whether or not there is an abnormality in the wire saw 1 based on the deviation information calculated by the deviation information calculation unit 54. In this embodiment, if the Mahalanobis distance as the deviation information does not exceed a preset threshold value, it is determined that there is no abnormality in the wire saw 1. Note that this determination is made separately for each of the first deviation information, the second deviation information, and the third deviation information that constitute the deviation information.

If it is determined that there is no abnormality in the wire saw 1 (No in step S5), proceed to step S7; if it is determined that there is an abnormality in the wire saw 1 (Yes in step S5), proceed to step S6.

In step S6, the factor estimation unit 56 estimates the cause of the abnormality based on the contribution of each of the multiple data items to the Mahalanobis distance as deviation information. For example, the factor estimation unit 56 estimates a data item that has a high contribution to increasing the Mahalanobis distance as a factor to be considered. In addition, more specific factors may be estimated, for example, empirically or experimentally, from the distribution of the contribution of each of the multiple data items.

In step S7, the result of the judgment by the judgment unit 55, i.e., the presence or absence of an abnormality in the wire saw 1, is displayed on the display device 26. Furthermore, if it is judged that there is an abnormality in the wire saw 1, the cause of the abnormality estimated by the cause estimation unit 56 is displayed on the display device 26. At this time, various information such as a warning message urging the user to stop operating the wire saw 1 and empirical or experimentally obtained countermeasures may be displayed on the display device 26.

As described above, the abnormality diagnosis device for the wire saw 1 according to this embodiment includes a diagnostic mode execution unit 51 that executes a first diagnostic mode in which the wire saw 1 is operated so that the processing roller 6 rotates at a constant first rotation speed before cutting, a second diagnostic mode in which the wire saw 1 is operated so that the pay-out reel 3 rotates at a constant second rotation speed, and a third diagnostic mode in which the wire saw 1 is operated so that the take-up reel 9 rotates at a constant third rotation speed, and a first data group obtained in the first diagnostic mode, a second data group obtained in the second diagnostic mode, and a third diagnostic mode in which the wire saw 1 is operated at a constant third rotation speed, for a plurality of data items indicating the operating state of the wire saw 1. The system includes a data group acquisition unit 52 that acquires a third data group obtained in each of the diagnostic modes, a deviation information calculation unit 54 that calculates deviation information regarding deviations obtained by comparing the first reference data group obtained in the first diagnostic mode when the wire saw 1 is normal, the second reference data group obtained in the second diagnostic mode when the wire saw 1 is normal, and the third reference data group obtained in the third diagnostic mode when the wire saw 1 is normal with the first data group, the second data group, and the third data group for the plurality of data items, respectively, and a determination unit 55 that determines whether or not there is an abnormality in the wire saw 1 based on the calculated deviation information.

In this embodiment, before cutting by the wire saw 1, a first diagnostic mode in which the processing roller 6 rotates at a constant speed, a second diagnostic mode in which the payout reel 3 rotates at a constant speed, and a third diagnostic mode in which the take-up reel 9 rotates at a constant speed are executed. Then, in each of the first to third diagnostic modes, first to third data groups are acquired for multiple data items, and the presence or absence of an abnormality in the wire saw 1 is determined based on deviation information calculated by comparing these with the first to third reference data groups under normal conditions.

The wire saw 1 is generally operated under a constant wire speed condition, and during such operation, the amount of wire stored and the effective diameter of the supply reel 3 and take-up reel 9 change each time. Furthermore, maintenance may cause the effective diameter of the processing roller 6 to change and decrease. In this way, the operating state before cutting is usually variable, and it was difficult to adequately monitor the operating state with conventional technology. In contrast, in this embodiment, before cutting, the first to third diagnostic modes are sequentially executed, in which the rotational speeds of the processing roller 6, supply reel 3, and take-up reel 9 are each set to a constant value independently, to acquire a group of data.

Therefore, according to this embodiment, the operating state of the wire saw 1 before the cutting process can be adequately monitored and an abnormality diagnosis can be performed accurately. This makes it possible to prevent an abnormality from occurring in the wire saw 1 during the cutting process, so that the wire saw 1 does not stop or require repair during the cutting process, leading to improved yield.

In addition, in this embodiment, the multiple data items used to obtain the first data group and the first reference data group include the torque load of the motor 12 that rotates the processing roller 6, the temperature of the processing roller 6 or the support member of the processing roller 6, and the vibration characteristics of the processing roller 6 or the support member of the processing roller 6. In this configuration, a data group is obtained for data items related to the processing roller 6 that are particularly relevant to abnormality diagnosis of the wire saw 1. This makes it possible to efficiently perform effective abnormality diagnosis.

In addition, in this embodiment, the multiple data items used to obtain the second data group and the second reference data group include the torque load of the motor 11 that rotates the payout reel 3, the temperature of the payout reel 3 or the support member of the payout reel 3, and the vibration characteristics of the payout reel 3 or the support member of the payout reel 3. In this configuration, a data group is obtained for data items related to the payout reel 3 that are highly relevant to abnormality diagnosis of the wire saw 1. This makes it possible to efficiently perform effective abnormality diagnosis.

In addition, in this embodiment, the multiple data items used to obtain the third data group and the third reference data group include the torque load of the motor 13 that rotates the take-up reel 9, the temperature of the take-up reel 9 or the support member of the take-up reel 9, and the vibration characteristics of the take-up reel 9 or the support member of the take-up reel 9. In this configuration, a data group is obtained for data items related to the take-up reel 9 that are highly relevant to abnormality diagnosis of the wire saw 1. This makes it possible to efficiently perform effective abnormality diagnosis.

In addition, in this embodiment, the deviation information calculation unit 54 calculates the Mahalanobis distance between a unit space formed by a preset reference data group and a signal space formed by an acquired data group as deviation information. In this configuration, the presence or absence of an abnormality in the wire saw 1 can be determined based on whether or not the Mahalanobis distance exceeds a predetermined threshold value.

This embodiment also includes a factor estimation unit 56 that estimates the cause of the abnormality based on the contribution of each of multiple data items to the deviation information. With this configuration, it becomes possible to take appropriate measures in the wire saw 1, such as replacing, repairing, or adjusting a part related to the cause of the estimated abnormality.

The present invention has been described above based on the embodiments, but the present invention is not limited to the configurations described in the above embodiments. The present invention can be modified in its configuration as appropriate without departing from the spirit of the invention, including by appropriately combining or selecting the configurations described in the above embodiments. In addition, parts of the configurations of the above embodiments can be added, deleted, or replaced.

For example, in the above embodiment, two processing rollers 6 are arranged, but this is not limited to this. For example, three rollers may be arranged at the vertices of an inverted triangle, or four rollers may be arranged at the vertices of a rectangle.

REFERENCE SIGNS LIST 1 wire saw 2 wire 3 pay-out reel 6 processing roller 9 take-up reel 11 to 13 motor 15 workpiece 31 to 33 vibration detection means 34 to 36 temperature detection means 51 diagnosis mode execution section 52 data group acquisition section 54 deviation information calculation section 55 judgment section 56 cause estimation section

Claims (7)

1. An abnormality diagnosis device for a wire saw that cuts a workpiece with a wire, the abnormality diagnosis device comprising: a plurality of processing rollers; a wire wound around the plurality of processing rollers; a pay-out reel that pays out the wire toward the processing rollers; and a take-up reel that takes up the wire fed from the processing rollers, the abnormality diagnosis device comprising:
a diagnostic mode execution unit that executes, before the cutting process, a first diagnostic mode in which the wire saw is operated so that the processing roller has a constant first rotation speed, a second diagnostic mode in which the wire saw is operated so that the pay-out reel has a constant second rotation speed, and a third diagnostic mode in which the wire saw is operated so that the take-up reel has a constant third rotation speed;
a data group acquiring unit that acquires a first data group obtained in the first diagnostic mode, a second data group obtained in the second diagnostic mode, and a third data group obtained in the third diagnostic mode, for a plurality of data items indicating an operating state of the wire saw;
a deviation information calculation unit that calculates deviation information regarding deviations between a first reference data group obtained in the first diagnostic mode when the wire saw is normal, a second reference data group obtained in the second diagnostic mode when the wire saw is normal, and a third reference data group obtained in the third diagnostic mode when the wire saw is normal, for a plurality of the data items, and the first data group, the second data group, and the third data group, respectively;
a determination unit that determines whether or not there is an abnormality in the wire saw based on the calculated deviation information; and
A wire saw abnormality diagnosis device comprising: The wire saw abnormality diagnosis device according to claim 1, characterized in that the multiple data items used to obtain the first data group and the first reference data group include the torque load of a motor that rotates and drives the processing roller, the temperature of the processing roller or the support member of the processing roller, and the vibration characteristics of the processing roller or the support member of the processing roller. The wire saw abnormality diagnosis device according to claim 1 or 2, characterized in that the multiple data items used to obtain the second data group and the second reference data group include the torque load of the motor that rotates the payout reel, the temperature of the payout reel or the support member of the payout reel, and the vibration characteristics of the payout reel or the support member of the payout reel. The wire saw abnormality diagnosis device according to any one of claims 1 to 3, characterized in that the multiple data items used to obtain the third data group and the third reference data group include the torque load of a motor that rotates and drives the take-up reel, the temperature of the take-up reel or a support member of the take-up reel, and the vibration characteristics of the take-up reel or a support member of the take-up reel. The wire saw abnormality diagnosis device according to any one of claims 1 to 4, characterized in that the deviation information calculation unit calculates, as the deviation information, Mahalanobis distances comparing a first unit space formed by the first reference data group, a second unit space formed by the second reference data group, and a third unit space formed by the third reference data group with a first signal space formed by the first data group, a second signal space formed by the second data group, and a third signal space formed by the third data group, respectively. The wire saw abnormality diagnosis device according to any one of claims 1 to 5, further comprising a factor estimation unit that, when it is determined that the wire saw has an abnormality, estimates the cause of the abnormality based on the contribution of each of the plurality of data items to the deviation information. A method for diagnosing an abnormality in a wire saw that cuts a workpiece with a wire, the method comprising: a plurality of processing rollers; a wire wound around the plurality of processing rollers; a pay-out reel that pays out the wire toward the processing rollers; and a take-up reel that takes up the wire fed from the processing rollers, the method comprising the steps of:
before the cutting process, a first diagnostic mode in which the wire saw is operated so that the processing roller rotates at a constant first rotation speed, a second diagnostic mode in which the wire saw is operated so that the pay-out reel rotates at a constant second rotation speed, and a third diagnostic mode in which the wire saw is operated so that the take-up reel rotates at a constant third rotation speed are respectively executed;
acquiring a first data group obtained in the first diagnostic mode, a second data group obtained in the second diagnostic mode, and a third data group obtained in the third diagnostic mode, for a plurality of data items indicating an operating state of the wire saw;
calculating deviation information regarding deviations between a first reference data group obtained in the first diagnostic mode when the wire saw is normal, a second reference data group obtained in the second diagnostic mode when the wire saw is normal, and a third reference data group obtained in the third diagnostic mode when the wire saw is normal, for a plurality of the data items, and comparing the first data group, the second data group, and the third data group, respectively;
determining whether or not there is an abnormality in the wire saw based on the calculated deviation information;
A method for diagnosing an abnormality in a wire saw, comprising:

Images